Re: Benchmarking Solr
Hi Benson, We typically use https://github.com/sematext/ActionGenerator As a matter of fact, we are using it right now to test one of our search clusters... Otis -- Solr & ElasticSearch Support http://sematext.com/ On Sun, May 26, 2013 at 10:38 AM, Benson Margulies wrote: > I'd like to run a repeatable test of having Solr ingest a corpus of > docs on disk, to measure the speed of some alternative things plugged > in. > > Anyone have some advice to share? One approach would be a quick SolrJ > program that pushed the entire stack as one giant collection with a > commit at the end.
Re: Benchmarking Solr
SolrMeter? Upayavira On Sun, May 26, 2013, at 03:38 PM, Benson Margulies wrote: > I'd like to run a repeatable test of having Solr ingest a corpus of > docs on disk, to measure the speed of some alternative things plugged > in. > > Anyone have some advice to share? One approach would be a quick SolrJ > program that pushed the entire stack as one giant collection with a > commit at the end.
Re: Benchmarking Solr 3.3 vs. 4.0
Shawn Heisey wrote: [..] For best results, you'll want to ensure that Solr4 is working completely from scratch, that it has never seen a 3.3 index, so that it will use its own native format. That's why I did in the second run. Thanks for clarifying that this is in fact better. :) It may be a good idea to look into the example Solr4 config/schema and see whether there are improvements you can make. One note: the updateLog feature in the update handler config will generally cause performance to be lower. The features that require updateLog would make this less of an apples to apples comparison, so I wouldn't enable it unless I knew I needed it. I'll have a look at the updateLog feature. But I'm pretty sure its disabled. Unless the lines are labelled wrong in the legend, the graph does show higher CPU usage during the push, but lower CPU usage during the optimize and most of the rest of the time. Slightly, but I was expecting higher latency also. Also raw data shows the box is unable to deliver CPU stats to the PerfMon Plugin because of high load. Perhaps I was expecting higher changes, but if you say what I see is ok, I'm fine. Can you comment on high CPU load even at low QPS rates? Is there some parameter to force lower load while testing at the cost of higher latencies for better comparison? The graph shows that Solr4 has lower latency than 3.3 during both the push and the optimize, as well as most of the rest of the time. The latency numbers however are a lot higher than I would expect, seeming to average out at around 100 seconds (10 ms). That is terrible performance from both versions. On my own Solr installation, which is distributed and has 78 million documents, I have a median latency of 8 milliseconds and a 95th percentile latency of 248 milliseconds. OK, I should relabel the y-axis because data is in fact 1000 times to high. So latency is more at 10ms which is quite good for high QPS rates. Is this a 64-bit platform with a 64-bit Java? How much memory have you allocated for the java heap? How big is the index? The VM I am using is an openSUSE 10.3 (i586), so no 64-bit Java at all (but production is using it). Tomcat Java parameters are: "-Xms1024m -Xmx1024m -XX:PermSize=256m -XX:MaxPermSize=256m -XX:+UseParallelGC -XX:ParallelGCThreads=4 -XX:GCTimeRatio=10" Number of docs is 266249 for both indices. Which is quite small, but I may be able to use a much larger index and a much better machine in the near future. Greetings Daniel Exner -- Daniel Exner Softwaredevelopment & Applicationsupport ESEMOS GmbH
Re: Benchmarking Solr 3.3 vs. 4.0
On 11/29/2012 8:29 AM, Daniel Exner wrote: I'll answer both your mails in one. Shawn Heisey wrote: On 11/29/2012 3:15 AM, Daniel Exner wrote: I'm currently doing some benchmarking of a real Solr 3.3 instance vs the same ported to Solr 4.0. [..] In the graph you can see high CPU load, all the time. This is even the case if I reduce the QPS down to 5, so CPU is no good metric for comparison between Solr 3.3 and 4.0 (at least on this machine). The missing memory data is due to the PerfMon JMeter Plugin having time-outs sometimes. You can also see no real increase in latency when pushing data into the index. This is puzzling me, as rumours say one should not push new data while under high load, as this would hurt query performance. I don't see any attachments, or any links to external attachments, so I can't see the graph. I can only make general statements, and I can't guarantee that they'll even be applicable to your scenario. You may need to use an external attachment service and just send us a link. Indeed, it seems like the mailing list daemon scrubbed my attachement. I dropped it into my Dropbox, here http://db.tt/EjYCqbpn Are you seeing lower performance, or just worried about the CPU load? Solr4 should be able to handle concurrent indexing and querying better than 3.x. It is able to do things concurrently that were not possible before. In general I'm interested in how much better Solr 4 performs and if it may be feasonable to use less powerful machines to get the same low latency, or do more data pushes etc. One way that performance improvements happen is that developers find slow sections of code where the CPU is fairly idle, and rewrite them so they are faster, but also exercise the CPU harder. When the new code runs, CPU load goes higher, but it all runs faster. Graphs show a slightly better latency for Solr 4.0 compared to 3.3, but not while pushing data. Another note specifically related to this part: Have you used the same configuration and done the minimal changes required to make it run, or have you tried to update the config for 4.0 and its considerable list of new features? Did you start with a blank index on 4.0, or did you copy the 3.3 index over? I used the same configuration and did the minimal changes. The first runs where using the same data from Solr 3.3 in Solr 4.0 (in fact it was even the same data dir..) but further runs used freshly filled different indices. For best results, you'll want to ensure that Solr4 is working completely from scratch, that it has never seen a 3.3 index, so that it will use its own native format. It may be a good idea to look into the example Solr4 config/schema and see whether there are improvements you can make. One note: the updateLog feature in the update handler config will generally cause performance to be lower. The features that require updateLog would make this less of an apples to apples comparison, so I wouldn't enable it unless I knew I needed it. Unless the lines are labelled wrong in the legend, the graph does show higher CPU usage during the push, but lower CPU usage during the optimize and most of the rest of the time. The graph shows that Solr4 has lower latency than 3.3 during both the push and the optimize, as well as most of the rest of the time. The latency numbers however are a lot higher than I would expect, seeming to average out at around 100 seconds (10 ms). That is terrible performance from both versions. On my own Solr installation, which is distributed and has 78 million documents, I have a median latency of 8 milliseconds and a 95th percentile latency of 248 milliseconds. Is this a 64-bit platform with a 64-bit Java? How much memory have you allocated for the java heap? How big is the index? Thanks, Shawn
Re: Benchmarking Solr 3.3 vs. 4.0
I'll answer both your mails in one. Shawn Heisey wrote: On 11/29/2012 3:15 AM, Daniel Exner wrote: I'm currently doing some benchmarking of a real Solr 3.3 instance vs the same ported to Solr 4.0. [..] In the graph you can see high CPU load, all the time. This is even the case if I reduce the QPS down to 5, so CPU is no good metric for comparison between Solr 3.3 and 4.0 (at least on this machine). The missing memory data is due to the PerfMon JMeter Plugin having time-outs sometimes. You can also see no real increase in latency when pushing data into the index. This is puzzling me, as rumours say one should not push new data while under high load, as this would hurt query performance. I don't see any attachments, or any links to external attachments, so I can't see the graph. I can only make general statements, and I can't guarantee that they'll even be applicable to your scenario. You may need to use an external attachment service and just send us a link. Indeed, it seems like the mailing list daemon scrubbed my attachement. I dropped it into my Dropbox, here http://db.tt/EjYCqbpn Are you seeing lower performance, or just worried about the CPU load? Solr4 should be able to handle concurrent indexing and querying better than 3.x. It is able to do things concurrently that were not possible before. In general I'm interested in how much better Solr 4 performs and if it may be feasonable to use less powerful machines to get the same low latency, or do more data pushes etc. One way that performance improvements happen is that developers find slow sections of code where the CPU is fairly idle, and rewrite them so they are faster, but also exercise the CPU harder. When the new code runs, CPU load goes higher, but it all runs faster. Graphs show a slightly better latency for Solr 4.0 compared to 3.3, but not while pushing data. Another note specifically related to this part: Have you used the same configuration and done the minimal changes required to make it run, or have you tried to update the config for 4.0 and its considerable list of new features? Did you start with a blank index on 4.0, or did you copy the 3.3 index over? I used the same configuration and did the minimal changes. The first runs where using the same data from Solr 3.3 in Solr 4.0 (in fact it was even the same data dir..) but further runs used freshly filled different indices. Greetings Daniel Exner -- Daniel Exner Softwaredevelopment & Applicationsupport ESEMOS GmbH
Re: Benchmarking Solr 3.3 vs. 4.0
On 11/29/2012 3:15 AM, Daniel Exner wrote: I'm currently doing some benchmarking of a real Solr 3.3 instance vs the same ported to Solr 4.0. Another note specifically related to this part: Have you used the same configuration and done the minimal changes required to make it run, or have you tried to update the config for 4.0 and its considerable list of new features? Did you start with a blank index on 4.0, or did you copy the 3.3 index over? There's no wrong answer to these questions. Depending on exactly what you are trying to do, what is right for someone else may not be right for you. The answers will help narrow the discussion. Thanks, Shawn
Re: Benchmarking Solr 3.3 vs. 4.0
On 11/29/2012 3:15 AM, Daniel Exner wrote: I'm currently doing some benchmarking of a real Solr 3.3 instance vs the same ported to Solr 4.0. Benchmarking is done using JMeter from localhost. Test scenario is a constant stream of queries from a log file out of production, at targeted 50 QPS. After some time (marked in graph) I do a push via REST interface of the whole index data (796M XML), wait some time and do a optimize via REST. Testmachine is a VM on a "Intel(R) Core(TM)2 Quad CPU Q9400 @2.66GH", one core and 2Gb RAM attached. Both Solr instances are running in the same Tomcat and are not used otherwise than testing. Expected results where a lower overall load for Solr 4 and a lower latency while pushing new data. In the graph you can see high CPU load, all the time. This is even the case if I reduce the QPS down to 5, so CPU is no good metric for comparison between Solr 3.3 and 4.0 (at least on this machine). The missing memory data is due to the PerfMon JMeter Plugin having time-outs sometimes. You can also see no real increase in latency when pushing data into the index. This is puzzling me, as rumours say one should not push new data while under high load, as this would hurt query performance. I don't see any attachments, or any links to external attachments, so I can't see the graph. I can only make general statements, and I can't guarantee that they'll even be applicable to your scenario. You may need to use an external attachment service and just send us a link. Are you seeing lower performance, or just worried about the CPU load? Solr4 should be able to handle concurrent indexing and querying better than 3.x. It is able to do things concurrently that were not possible before. One way that performance improvements happen is that developers find slow sections of code where the CPU is fairly idle, and rewrite them so they are faster, but also exercise the CPU harder. When the new code runs, CPU load goes higher, but it all runs faster. Thanks, Shawn
RE: Benchmarking Solr
Hi, I agree with Jean-Sebastien. JMeter is great! The threads in my test plan are configured to use an "Access Log Sampler". This allows you to feed your production requests through JMeter, simulating production traffic. When I launch the test, it has access to about 3 million production queries. I also use the "user defined variables" in my test plan so I can customize different parameters at runtime. http://jakarta.apache.org/jmeter/usermanual/jmeter_accesslog_sampler_step_by_step.pdf - Erica -Original Message- From: Jean-Sebastien Vachon [mailto:js.vac...@videotron.ca] Sent: Saturday, April 10, 2010 11:04 PM To: solr-user@lucene.apache.org Subject: Re: Benchmarking Solr Hi, why don't you use JMeter? It would give you greater control over the tests you wish to make. It has many different samplers that will let you run different scenarios using your existing set of queries. ab is great when you want to evaluate the performance of your server under heavy load. But other than this, I don`t see much use to it. JMeter offers many more options once you get to know it a little. good luck - Original Message - From: "Blargy" To: Sent: Friday, April 09, 2010 9:46 PM Subject: Benchmarking Solr > > I am about to deploy Solr into our production environment and I would like > to > do some benchmarking to determine how many slaves I will need to set up. > Currently the only way I know how to benchmark is to use Apache Benchmark > but I would like to be able to send random requests to the Solr... not > just > one request over and over. > > I have a sample data set of 5000 user entered queries and I would like to > be > able to use AB to benchmark against all these random queries. Is this > possible? > > FYI our current index is ~1.5 gigs with ~5m documents and we will be using > faceting quite extensively. Are average requests per/day is ~2m. We will > be > running RHEL with about 8-12g ram. Any idea how many slaves might be > required to handle our load? > > Thanks > -- > View this message in context: > http://n3.nabble.com/Benchmarking-Solr-tp709561p709561.html > Sent from the Solr - User mailing list archive at Nabble.com.
Re: Benchmarking Solr
On 4/12/2010 9:57 AM, Shawn Heisey wrote: On 4/12/2010 8:51 AM, Paolo Castagna wrote: There are already two related pages: - http://wiki.apache.org/solr/SolrPerformanceFactors - http://wiki.apache.org/solr/SolrPerformanceData Why not to create a new page? - http://wiki.apache.org/solr/BenchmarkingSolr (?) Done. I hope you like it. Please feel free to improve it. http://wiki.apache.org/solr/BenchmarkingSolr I've updated the script to clean things up further, but the most important change is that it will now do URI escaping on the query string before plugging it into the URL. You can turn this off if your query strings are already escaped.
Re: Benchmarking Solr
On 4/12/2010 8:51 AM, Paolo Castagna wrote: There are already two related pages: - http://wiki.apache.org/solr/SolrPerformanceFactors - http://wiki.apache.org/solr/SolrPerformanceData Why not to create a new page? - http://wiki.apache.org/solr/BenchmarkingSolr (?) Done. I hope you like it. Please feel free to improve it. http://wiki.apache.org/solr/BenchmarkingSolr
Re: Benchmarking Solr
Shawn Heisey wrote: Anyone got a recommendation about where to put it on the wiki? There are already two related pages: - http://wiki.apache.org/solr/SolrPerformanceFactors - http://wiki.apache.org/solr/SolrPerformanceData Why not to create a new page? - http://wiki.apache.org/solr/BenchmarkingSolr (?) It would be good to have someone using JMeter to share their config files as well. Paolo
Re: Benchmarking Solr
Paolo Castagna wrote: I do not have an answer to your questions. But, I have the same issue/problem you have. Some related threads: - http://markmail.org/message/pns4dtfvt54mu3vs - http://markmail.org/message/7on6lvabsosvj7bc - http://markmail.org/message/ftz7tkd7ekhnk4bc - http://markmail.org/message/db2cv3dzakdp23qm - http://markmail.org/message/m3x6ogkfdhcwae6z - http://markmail.org/message/xoe3ny7dldnx4wby - http://markmail.org/message/eoqty4ralk34rgzk Paolo
RE: Benchmarking Solr
I have been using Jmeter to perform some load testing. In your case you might like to take a look at http://jakarta.apache.org/jmeter/usermanual/component_reference.html#CSV_Data_Set_Config . This will allow you to use a random item from your query list. Regards, Kallin Nagelberg -Original Message- From: Blargy [mailto:zman...@hotmail.com] Sent: Friday, April 09, 2010 9:47 PM To: solr-user@lucene.apache.org Subject: Benchmarking Solr I am about to deploy Solr into our production environment and I would like to do some benchmarking to determine how many slaves I will need to set up. Currently the only way I know how to benchmark is to use Apache Benchmark but I would like to be able to send random requests to the Solr... not just one request over and over. I have a sample data set of 5000 user entered queries and I would like to be able to use AB to benchmark against all these random queries. Is this possible? FYI our current index is ~1.5 gigs with ~5m documents and we will be using faceting quite extensively. Are average requests per/day is ~2m. We will be running RHEL with about 8-12g ram. Any idea how many slaves might be required to handle our load? Thanks -- View this message in context: http://n3.nabble.com/Benchmarking-Solr-tp709561p709561.html Sent from the Solr - User mailing list archive at Nabble.com.
Re: Benchmarking Solr
I've got a very simple perl script (most of the work is done with modules) that I wrote which forks off multiple processes and throws requests at Solr, then gives a little bit of statistical analysis at the end. I have planned on sharing it from the beginning, I just have to clean it up for public consumption. I will try to do that today, though I don't know if I can. Anyone got a recommendation about where to put it on the wiki? Shawn On 4/9/2010 7:46 PM, Blargy wrote: I am about to deploy Solr into our production environment and I would like to do some benchmarking to determine how many slaves I will need to set up. Currently the only way I know how to benchmark is to use Apache Benchmark but I would like to be able to send random requests to the Solr... not just one request over and over. I have a sample data set of 5000 user entered queries and I would like to be able to use AB to benchmark against all these random queries. Is this possible? FYI our current index is ~1.5 gigs with ~5m documents and we will be using faceting quite extensively. Are average requests per/day is ~2m. We will be running RHEL with about 8-12g ram. Any idea how many slaves might be required to handle our load? Thanks
Re: Benchmarking Solr
Hi, You can use Siege [1] in a similar manner as AB and it can support newline separated URL files and pick random URL's. [1]:http://freshmeat.net/projects/siege/ Cheers, On Saturday 10 April 2010 03:46:56 Blargy wrote: > I am about to deploy Solr into our production environment and I would like > to do some benchmarking to determine how many slaves I will need to set > up. Currently the only way I know how to benchmark is to use Apache > Benchmark but I would like to be able to send random requests to the > Solr... not just one request over and over. > > I have a sample data set of 5000 user entered queries and I would like to > be able to use AB to benchmark against all these random queries. Is this > possible? > > FYI our current index is ~1.5 gigs with ~5m documents and we will be using > faceting quite extensively. Are average requests per/day is ~2m. We will be > running RHEL with about 8-12g ram. Any idea how many slaves might be > required to handle our load? > > Thanks > Markus Jelsma - Technisch Architect - Buyways BV http://www.linkedin.com/in/markus17 050-8536620 / 06-50258350
Re: Benchmarking Solr
Hi, I do not have an answer to your questions. But, I have the same issue/problem you have. It would be good if Solr community would agree and share their approach for benchmarking Solr. Indeed, it would be good to have a benchmark for "information retrieval" systems. AFIK there isn't one. :-/ The content on the wiki [1] is better than nothing, but in practice more is needed IMHO. I have seen JMeter being used in ElasticSearch [2]. Solr could do the same to help users and new adopters to start. Some guidelines/advices (I know it's hard) would be useful as well. I ended up writing my own "crappy" multi-threaded benchmarking tool. Also, are you using Jetty? At a certain point, in particular when you are hitting the Solr cache and returning a large number of results, the transfer time is a significant part of your response time. Tuning Jetty or Tomcat or something else is essential. Are you using Jetty or Tomcat? I would also be interested in understanding the impact of the slave pooling interval on searches and the impact of the number of slaves and pooling interval on updates on the master. Paolo [1] http://wiki.apache.org/solr/SolrPerformanceData [2] http://github.com/elasticsearch/elasticsearch/tree/master/modules/benchmark/jmeter Blargy wrote: I am about to deploy Solr into our production environment and I would like to do some benchmarking to determine how many slaves I will need to set up. Currently the only way I know how to benchmark is to use Apache Benchmark but I would like to be able to send random requests to the Solr... not just one request over and over. I have a sample data set of 5000 user entered queries and I would like to be able to use AB to benchmark against all these random queries. Is this possible? FYI our current index is ~1.5 gigs with ~5m documents and we will be using faceting quite extensively. Are average requests per/day is ~2m. We will be running RHEL with about 8-12g ram. Any idea how many slaves might be required to handle our load? Thanks
Re: Benchmarking Solr
Hi, why don't you use JMeter? It would give you greater control over the tests you wish to make. It has many different samplers that will let you run different scenarios using your existing set of queries. ab is great when you want to evaluate the performance of your server under heavy load. But other than this, I don`t see much use to it. JMeter offers many more options once you get to know it a little. good luck - Original Message - From: "Blargy" To: Sent: Friday, April 09, 2010 9:46 PM Subject: Benchmarking Solr I am about to deploy Solr into our production environment and I would like to do some benchmarking to determine how many slaves I will need to set up. Currently the only way I know how to benchmark is to use Apache Benchmark but I would like to be able to send random requests to the Solr... not just one request over and over. I have a sample data set of 5000 user entered queries and I would like to be able to use AB to benchmark against all these random queries. Is this possible? FYI our current index is ~1.5 gigs with ~5m documents and we will be using faceting quite extensively. Are average requests per/day is ~2m. We will be running RHEL with about 8-12g ram. Any idea how many slaves might be required to handle our load? Thanks -- View this message in context: http://n3.nabble.com/Benchmarking-Solr-tp709561p709561.html Sent from the Solr - User mailing list archive at Nabble.com.